A new class of electron microscopy labels is being developed that has 5-10 times better resolution than those currently available. This class promises to be the next generation of high resolution electron microscopic labels and should have an extensive impact on molecular biology, since it should make possible direct domainal mapping of single biomolecules and macromolecular assemblies. Previously we were able to introduce undecagold (Au11) as a specifically bound protein label. We also functionalized it in several ways to demonstrate its utility in labeling sulfhydryls, amino groups and carbohydrate residues in glycoproteins. Pioneering work in this new field will continue along 3 lines: 1) Synthetic chemistry to produce a wider range of labels with varying properties and reactive groups. 2) Basic physical studies to properly characterize these labels covering their visibility, detection limits and sensitivity to the electron beam. 3) Biological applications to demonstrate the variety of labeling available, its reliability, as well as answering some important biological structural questions not easily possible with any other techniques. Chemical synthesis and characterization will focus on development of a monofunctional undecatungsten (W11) label that has just recently been successfully made for the first time. This W11 label (with organic side chain) has several advantages over the Au11 label and will be derivatized to react with various protein, nucleic acid and carbohydrate moieties. Biological applications will include labeling of: t-RNA and ribosome components, mutienzyme complexes (pyruvate dehydrogenase complex), actin, myosin, intermediate filaments, human fibrinogen, acetytcholine receptors and membranes. Fab fragments will also be labeled to provide an antibody label 5 times smaller than those currently available. A new electron gun allowing low dose/high dose data to be collected with no optical changes will enable minimally damaged specimens to be used. A stable cold stage will permit frozen hydrated cluster-labeled samples to be observed. Image processing to optimally recover high resolution data will also be employed especially in radial and helical reconstructions. We will draw upon our large user program of the STEM NIH Biotechnology Resource to provide interesting samples and biochemical expertise since many users are very anxious to try these new labels.